Skip to content

Actions: ngxson/llama.cpp

Actions

Server

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
server.yml
900 workflow run results
900 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

WIP Pixtral Server #1357: Pull request #22 opened by ngxson
8m 28s xsn/mtmd_pixtral
xsn/mtmd_pixtral
8m 28s
ggml : add SSE 4.2 and x64 base variant for CPUs without AVX (#12871) Server #1356: Commit 1d735c0 pushed by ngxson
8m 48s master
master
8m 48s
SYCL: Add non-contiguous support in ROPE (#12993) Server #1355: Commit 5368ddd pushed by ngxson
12m 45s master
master
12m 45s
mtmd : merge llava, gemma3 and minicpmv CLI into single `llama-mtmd-c… Server #1354: Commit 84a9bf2 pushed by ngxson
9m 33s master
master
9m 33s
convert : experimental support for --mmproj flag (#13023) Server #1353: Commit 2016f07 pushed by ngxson
13m 14s master
master
13m 14s
llava: fix errors in clip.h on certain compilers (#13030) Server #1352: Commit 6602304 pushed by ngxson
8m 17s master
master
8m 17s
vulkan: support noncontiguous rms_norm (#13031) Server #1351: Commit 6616820 pushed by ngxson
9m 56s master
master
9m 56s
metal: add neg operator (#13029) Server #1350: Commit 4ba9d71 pushed by ngxson
8m 23s master
master
8m 23s
clip : refactor, add image_manipulation and llava_uhd classes (#1… Server #1349: Commit 37b9f0d pushed by ngxson
9m 31s master
master
9m 31s
main : Fix Ctrl+D/newline handling (#12951) Server #1348: Commit 6408210 pushed by ngxson
9m 47s master
master
9m 47s
server : use std::move whenever possible (#12936) Server #1347: Commit 35370ba pushed by ngxson
8m 27s master
master
8m 27s
SYCL: Refactor and enable FP16 in binary broadcast OPs (#12975) Server #1346: Commit 8d66005 pushed by ngxson
9m 0s master
master
9m 0s
mtmd : add methods to access mtmd_image_tokens (#12906) Server #1345: Commit b9154ec pushed by ngxson
8m 36s master
master
8m 36s
rpc : add RPC_CMD_HELLO (#12955) Server #1344: Commit 2db9ba1 pushed by ngxson
8m 31s master
master
8m 31s
graph : make FA compatible with MLA + add initial Metal kernels (#12953) Server #1343: Commit 2f74c35 pushed by ngxson
9m 28s master
master
9m 28s
ggml: Re-enable CUDA graphs in presence of CONT and DUP nodes (#12970) Server #1342: Commit 207c22e pushed by ngxson
9m 22s master
master
9m 22s
CANN: Add support for async operator submission (#12864) Server #1341: Commit 7a395f6 pushed by ngxson
9m 19s master
master
9m 19s
llama : recognize IBM Granite 3.3 FIM tokens (#12988) Server #1340: Commit 971f245 pushed by ngxson
11m 55s master
master
11m 55s
opencl: fix incorrect local_size index in profiling log (#12868) Server #1339: Commit 12b1750 pushed by ngxson
8m 43s master
master
8m 43s
vulkan: enable coopmat2 FA gqa and split_k optimizations more often (… Server #1338: Commit 015022b pushed by ngxson
8m 59s master
master
8m 59s
CANN: Add 310P operator support check (#12962) Server #1337: Commit b43d89e pushed by ngxson
10m 52s master
master
10m 52s
opencl: split ggml-opencl.cl into multiple files and cleanup (#12886) Server #1336: Commit 80f19b4 pushed by ngxson
8m 25s master
master
8m 25s
metal : add FA-vec kernels for head size 96 (#12952) Server #1335: Commit f8f820c pushed by ngxson
8m 36s master
master
8m 36s
CANN: Add x86 build ci (#12950) Server #1334: Commit 54a7272 pushed by ngxson
8m 24s master
master
8m 24s
CUDA/HIP: Share the same unified memory allocation logic. (#12934) Server #1333: Commit 84778e9 pushed by ngxson
6m 38s master
master
6m 38s